Seizure recurrence and intractability cause significant morbidity and mortality in patients with epilepsy, as well as significant individual and societal costs. Antiepileptic drug (AED) therapy often follows a course of remission and relapse, and many patients do not achieve seizure freedom. What increases patient's risk for seizure recurrence remains incompletely understood, except for specific risk factors. Critically, the pathologic process that ultimately leads to epilepsy likely begins well before the clinical onset o the disease, making early intervention attractive. The functional organization of the brain of patients with late stage epilepsy is fundamentally altered. The long-term goal of this research is to use functional neuroimaging techniques to identify patterns of brain function reorganization in early epilepsy (and ideally preclinical epilepsy) that predict future disease severity and can be used by physicians to guide early intervention and more aggressive therapy. The central hypothesis is that the functional reorganization of epileptic brains near the clinical onset of the disease reflects the underlying epileptogenic process, including an individual's propensity for seizure recurrence, while changes that occur in later stages of the disease reflect disease progression/seizure burden. This project has thus two objectives. The first objective is for the candidate to use a prospective, longitudinal study of adults to assess functional network changes in early epilepsy and determine whether they can predict seizure recurrence or intractability. The second objective is to leverage the extremely supportive environment at Washington University to provide the candidate with training in 1) methods of longitudinal study design and conduction and 2) methods of multivariate statistics applied to multidimensional systems, such as those studied with functional connectivity. The goal is to support the development of an R01 application and a progression to independence with a research program that uses advanced neuroimaging techniques to study epilepsy prospectively, beginning in the early stages of the disease. Findings from this project would have great applicability to clinical practice. Specifically the ability to predict disease severity at an early stage using abnormal functional network properties could significantly reduce morbidity and mortality by leading clinicians to pursue more aggressive AED therapy earlier in the course, and by allowing tracking of disease progression and early consideration of other therapeutic options, including epilepsy surgery. Furthermore the data obtained with this proposal will be the basis of a larger prospective longitudinal study in the functional network changes of early epilepsy, with the goal of translating this knowledge into individually tailored interventions targeting the disease before it produces irreversible damage.